Universal Transverse Mercator coordinate system

Simplified view of contiguous US UTM zones, projected with Lambert conformal conic.

The UTM system divides the Earth into 60 zones, each 6° of longitude in width. Zone 1 covers longitude 180° to 174° W; zone numbering increases eastward to zone 60, which covers longitude 174°E to 180°. The polar regions south of 80°S and north of 84°N are excluded.

Each of the 60 zones uses a transverse Mercator projection that can map a region of large north-south extent with low distortion. By using narrow zones of 6° of longitude (up to 668 km) in width, and reducing the scale factor along the central meridian to 0.9996 (a reduction of 1:2500), the amount of distortion is held below 1 part in 1,000 inside each zone. Distortion of scale increases to 1.0010 at the zone boundaries along the equator.

In each zone the scale factor of the central meridian reduces the diameter of the transverse cylinder to produce a secant projection with two standard lines, or lines of true scale, about 180 km on each side of, and about parallel to, the central meridian (Arc cos 0.9996 = 1.62° at the Equator). The scale is less than 1 inside the standard lines and greater than 1 outside them, but the overall distortion is minimized.

Universal Transverse Mercator (UTM) Grid Zones 31N thru 37N differ from the standard 6° wide by 84° zone for the northern hemisphere, in part to accommodate the southern half of the Kingdom of Norway. For more on its history, see Clifford J. Mugnier's article on Grids & Datums of The Kingdom of Norway that appeared in the October 1999 issue of PE&RS http://www.asprs.org/a/resources/grids/10-99-norway.pdf

Distortion of scale increases in each UTM zone as the boundaries between the UTM zones are approached. However, it is often convenient or necessary to measure a series of locations on a single grid when some are located in two adjacent zones. Around the boundaries of large scale maps (1:100,000 or larger) coordinates for both adjoining UTM zones are usually printed within a minimum distance of 40 km on either side of a zone boundary. Ideally, the coordinates of each position should be measured on the grid for the zone in which they are located, but because the scale factor is still relatively small near zone boundaries, it is possible to overlap measurements into an adjoining zone for some distance when necessary.

Each zone is segmented into 20 latitude bands. Each latitude band is 8 degrees high, and is lettered starting from "C" at 80°S, increasing up the English alphabet until "X", omitting the letters "I" and "O" (because of their similarity to the numerals one and zero). The last latitude band, "X", is extended an extra 4 degrees, so it ends at 84°N latitude, thus covering the northernmost land on Earth.

Latitude bands "A" and "B" do exist, as do bands "Y" and "Z". They cover the western and eastern sides of the Antarctic and Arctic regions respectively. A convenient mnemonic to remember is that the letter "N" is the first letter in "northern hemisphere", so any letter coming before "N" in the alphabet is in the southern hemisphere, and any letter "N" or after is in the northern hemisphere.

The combination of a zone and a latitude band defines a grid zone. The zone is always written first, followed by the latitude band. For example, (see image, top right), a position in Toronto, Ontario, Canada, would find itself in zone 17 and latitude band "T", thus the full grid zone reference is "17T". The grid zones serve to delineate irregular UTM zone boundaries. They also are an integral part of the military grid reference system.

A note of caution: A method also is used that simply adds N or S following the zone number to indicate North or South hemisphere (the easting and northing coordinates along with the zone number supplying everything necessary to geolocate a position except which hemisphere). However, this method has caused some confusion since, for instance, "50S" can mean southern hemisphere but also grid zone "50S" in the northern hemisphere.[8] There are many possible ways to disambiguate between the two methods, two of which are demonstrated later in this article.

These grid zones are uniform over the globe, except in two areas. On the southwest coast of Norway, grid zone 32V (9° of longitude in width) is extended further west, and grid zone 31V (3° of longitude in width) is correspondingly shrunk to cover only open water. Also, in the region around Svalbard, the four grid zones 31X (9° of longitude in width), 33X (12° of longitude in width), 35X (12° of longitude in width), and 37X (9° of longitude in width) are extended to cover what would otherwise have been covered by the seven grid zones 31X to 37X. The three grid zones 32X, 34X and 36X are not used.

• Grid zones in various parts of the world
• 
  Europe
• 
  Africa
• 
  South America
• 
  Bering Sea with Alaska

These formulae are truncated version of Transverse Mercator: flattening series, which were originally derived by Johann Heinrich Louis Krüger in 1912.[9] They are accurate to around a millimeter within 3,000 km of the central meridian.[10] Concise commentaries for their derivation have also been given.[11][12]

The WGS 84 spatial reference system describes Earth as an oblate spheroid along north-south axis with an equatorial radius of ${\displaystyle a=6378.137}$ km and an inverse flattening of ${\displaystyle 1/f=298.257\,223\,563}$. Let's take a point of latitude ${\displaystyle \,\varphi }$ and of longitude ${\displaystyle \,\lambda }$ and compute its UTM coordinates as well as point scale factor ${\displaystyle k\,\!}$ and meridian convergence ${\displaystyle \gamma \,\!}$ using a reference meridian of longitude ${\displaystyle \lambda _{0}}$. By convention, in the northern hemisphere ${\displaystyle N_{0}=0}$ km and in the southern hemisphere ${\displaystyle N_{0}=10000}$ km. By convention also ${\displaystyle k_{0}=0.9996}$ and ${\displaystyle E_{0}=500}$ km.

In the following formulas, the distances are in kilometers. In advance let's compute some preliminary values:

${\displaystyle n={\frac {f}{2-f}},\quad A={\frac {a}{1+n}}\left(1+{\frac {n^{2}}{4}}+{\frac {n^{4}}{64}}+\cdots \right),}$

${\displaystyle \alpha _{1}={\frac {1}{2}}n-{\frac {2}{3}}n^{2}+{\frac {5}{16}}n^{3},\,\,\,\alpha _{2}={\frac {13}{48}}n^{2}-{\frac {3}{5}}n^{3},\,\,\,\alpha _{3}={\frac {61}{240}}n^{3},}$

${\displaystyle \beta _{1}={\frac {1}{2}}n-{\frac {2}{3}}n^{2}+{\frac {37}{96}}n^{3},\,\,\,\beta _{2}={\frac {1}{48}}n^{2}+{\frac {1}{15}}n^{3},\,\,\,\beta _{3}={\frac {17}{480}}n^{3},}$

${\displaystyle \delta _{1}=2n-{\frac {2}{3}}n^{2}-2n^{3},\,\,\,\delta _{2}={\frac {7}{3}}n^{2}-{\frac {8}{5}}n^{3},\,\,\,\delta _{3}={\frac {56}{15}}n^{3}.}$

First let's compute some intermediate values:

${\displaystyle t=\sinh \left(\tanh ^{-1}\left(\sin \varphi \right)-{\frac {2{\sqrt {n}}}{1+n}}\tanh ^{-1}\left({\frac {2{\sqrt {n}}}{1+n}}\sin \varphi \right)\right),}$

${\displaystyle \xi '=\tan ^{-1}\left({\frac {t}{\cos(\lambda -\lambda _{0})}}\right),\,\,\,\eta '=\tanh ^{-1}\left({\frac {\sin(\lambda -\lambda _{0})}{\sqrt {1+t^{2}}}}\right),}$

${\displaystyle \sigma =1+\sum _{j=1}^{3}2j\alpha _{j}\cos(2j\xi ')\cosh(2j\eta '),\,\,\,\tau =\sum _{j=1}^{3}2j\alpha _{j}\sin(2j\xi ')\sinh(2j\eta ').}$

The final formulae are:

${\displaystyle E=E_{0}+k_{0}A\left(\eta '+\sum _{j=1}^{3}\alpha _{j}\cos(2j\xi ')\sinh(2j\eta ')\right),}$

${\displaystyle N=N_{0}+k_{0}A\left(\xi '+\sum _{j=1}^{3}\alpha _{j}\sin(2j\xi ')\cosh(2j\eta ')\right),}$

${\displaystyle k={\frac {k_{0}A}{a}}{\sqrt {\left\{1+\left({\frac {1-n}{1+n}}\tan \varphi \right)^{2}\right\}{\frac {\sigma ^{2}+\tau ^{2}}{t^{2}+\cos ^{2}(\lambda -\lambda _{0})}}}},}$

${\displaystyle \gamma =\tan ^{-1}\left({\frac {\tau {\sqrt {1+t^{2}}}+\sigma t\tan(\lambda -\lambda _{0})}{\sigma {\sqrt {1+t^{2}}}-\tau t\tan(\lambda -\lambda _{0})}}\right).}$

where ${\displaystyle E}$ is Easting, ${\displaystyle N}$ is Northing, ${\displaystyle k}$ is the Scale Factor, and ${\displaystyle \gamma }$ is the Grid Convergence.

Note: Hemi=+1 for Northern, Hemi=-1 for Southern

First let's compute some intermediate values:

${\displaystyle \xi ={\frac {N-N_{0}}{k_{0}A}},\,\,\,\eta ={\frac {E-E_{0}}{k_{0}A}},}$

${\displaystyle \xi '=\xi -\sum _{j=1}^{3}\beta _{j}\sin \left(2j\xi \right)\cosh \left(2j\eta \right),\,\,\,\eta '=\eta -\sum _{j=1}^{3}\beta _{j}\cos \left(2j\xi \right)\sinh \left(2j\eta \right),}$

${\displaystyle \sigma '=1-\sum _{j=1}^{3}2j\beta _{j}\cos \left(2j\xi \right)\cosh \left(2j\eta \right),\,\,\,\tau '=\sum _{j=1}^{3}2j\beta _{j}\sin \left(2j\xi \right)\sinh \left(2j\eta \right),}$

${\displaystyle \chi =\sin ^{-1}\left({\frac {\sin \xi '}{\cosh \eta '}}\right).}$

The final formulae are:

${\displaystyle \varphi =\chi +\sum _{j=1}^{3}\delta _{j}\sin \left(2j\chi \right),}$

${\displaystyle \lambda _{0}=\mathrm {Z} \mathrm {o} \mathrm {n} \mathrm {e} \times 6^{\circ }-183^{\circ }\,}$

${\displaystyle \lambda =\lambda _{0}+\tan ^{-1}\left({\frac {\sinh \eta '}{\cos \xi '}}\right),}$

${\displaystyle k={\frac {k_{0}A}{a}}{\sqrt {\left\{1+\left({\frac {1-n}{1+n}}\tan \varphi \right)^{2}\right\}{\frac {\cos ^{2}\xi '+\sinh ^{2}\eta '}{\sigma '^{2}+\tau '^{2}}}}},}$

${\displaystyle \gamma =\mathrm {H} \mathrm {e} \mathrm {m} \mathrm {i} \times \tan ^{-1}\left({\frac {\tau '+\sigma '\tan \xi '\tanh \eta '}{\sigma '-\tau '\tan \xi '\tanh \eta '}}\right).}$